Data Engineer

Minerva · Warsaw

Minerva runs on data most companies never see in one place: tenders from dozens of procurement platforms, in multiple languages, in formats that were clearly designed to be printed. Our AI is only as good as the pipelines feeding it. That's the job. The data layer is our moat. Every feature we ship stands on it, and right now the people building it can count themselves on one hand. You'll make architecture decisions that stick, see your work in front of customers within days, and shape how a company that lives on data actually handles it. Responsibilities Build and own the pipelines that ingest tender data from public procurement sources across Poland and the EU: announcements, specifications, attachments, amendments, results. Parse documents that are hostile by default: scanned PDFs, ZIPs inside ZIPs, tables that aren't tables. Keep the data fresh, deduplicated, and structured so the product and the AI layer can rely on it. Monitor sources that change without warning and build systems that notice before customers do. Work directly with product engineers on what the data model needs to support next: new markets, new document types, new features. Requirements Strong Python and SQL, with production data pipelines you've built and maintained: scraping, ETL, orchestration, monitoring. You treat data quality as an engineering problem, not a cleanup task. Comfortable owning a system end to end: if it breaks at the source, you find out why. Experience with document processing, OCR, or LLM pipelines is a plus. Any exposure to public procurement data helps. But we'll teach you the domain faster than you'd expect. What we offer Competitive compensation Multisport card Transparent comp: real numbers on the call, written down, no grey areas

More roles at Minerva